On Fri, Mar 08, 2013 at 04:46:15AM +0100, Marc Zyngier wrote: > On Thu, 7 Mar 2013 10:09:03 +0200, "Michael S. Tsirkin" <mst@xxxxxxxxxx> > wrote: > > On Tue, Mar 05, 2013 at 03:47:24AM +0000, Marc Zyngier wrote: > >> Provide the kvm.h file that defines the user space visible > >> interface. > >> > >> Signed-off-by: Marc Zyngier <marc.zyngier@xxxxxxx> > >> --- > >> arch/arm64/include/uapi/asm/kvm.h | 112 > >> ++++++++++++++++++++++++++++++++++++++ > >> 1 file changed, 112 insertions(+) > >> create mode 100644 arch/arm64/include/uapi/asm/kvm.h > >> > >> diff --git a/arch/arm64/include/uapi/asm/kvm.h > >> b/arch/arm64/include/uapi/asm/kvm.h > >> new file mode 100644 > >> index 0000000..f5525f1 > >> --- /dev/null > >> +++ b/arch/arm64/include/uapi/asm/kvm.h > >> @@ -0,0 +1,112 @@ > >> +/* > >> + * Copyright (C) 2012 - ARM Ltd > >> + * Author: Marc Zyngier <marc.zyngier@xxxxxxx> > >> + * > >> + * Derived from arch/arm/include/uapi/asm/kvm.h: > >> + * Copyright (C) 2012 - Virtual Open Systems and Columbia University > >> + * Author: Christoffer Dall <c.dall@xxxxxxxxxxxxxxxxxxxxxx> > >> + * > >> + * This program is free software; you can redistribute it and/or > modify > >> + * it under the terms of the GNU General Public License version 2 as > >> + * published by the Free Software Foundation. > >> + * > >> + * This program is distributed in the hope that it will be useful, > >> + * but WITHOUT ANY WARRANTY; without even the implied warranty of > >> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the > >> + * GNU General Public License for more details. > >> + * > >> + * You should have received a copy of the GNU General Public License > >> + * along with this program. If not, see > <http://www.gnu.org/licenses/>. > >> + */ > >> + > >> +#ifndef __ARM_KVM_H__ > >> +#define __ARM_KVM_H__ > >> + > >> +#define KVM_SPSR_EL1 0 > >> +#define KVM_NR_SPSR 1 > >> + > >> +#ifndef __ASSEMBLY__ > >> +#include <asm/types.h> > >> +#include <asm/ptrace.h> > >> + > >> +#define __KVM_HAVE_GUEST_DEBUG > >> +#define __KVM_HAVE_IRQ_LINE > >> + > >> +#define KVM_REG_SIZE(id) \ > >> + (1U << (((id) & KVM_REG_SIZE_MASK) >> KVM_REG_SIZE_SHIFT)) > >> + > >> +struct kvm_regs { > >> + struct user_pt_regs regs; /* sp = sp_el0 */ > >> + > >> + unsigned long sp_el1; > >> + unsigned long elr_el1; > >> + > >> + unsigned long spsr[KVM_NR_SPSR]; > >> +}; > >> + > > > > Using long in uapi is generally a mistake: with gcc it has > > different size depending on whether you build in 64 and 32 bit mode. > > I think it is better to make it __u64. > > Ah, well spotted. Definitely a mistake. > > >> +/* Supported Processor Types */ > >> +#define KVM_ARM_TARGET_CORTEX_A57 0 > >> +#define KVM_ARM_NUM_TARGETS 1 > >> + > >> +/* KVM_ARM_SET_DEVICE_ADDR ioctl id encoding */ > >> +#define KVM_ARM_DEVICE_TYPE_SHIFT 0 > >> +#define KVM_ARM_DEVICE_TYPE_MASK (0xffff << KVM_ARM_DEVICE_TYPE_SHIFT) > >> +#define KVM_ARM_DEVICE_ID_SHIFT 16 > >> +#define KVM_ARM_DEVICE_ID_MASK (0xffff << KVM_ARM_DEVICE_ID_SHIFT) > >> + > >> +/* Supported device IDs */ > >> +#define KVM_ARM_DEVICE_VGIC_V2 0 > >> + > >> +/* Supported VGIC address types */ > >> +#define KVM_VGIC_V2_ADDR_TYPE_DIST 0 > >> +#define KVM_VGIC_V2_ADDR_TYPE_CPU 1 > >> + > >> +#define KVM_VGIC_V2_DIST_SIZE 0x1000 > >> +#define KVM_VGIC_V2_CPU_SIZE 0x2000 > >> + > >> +struct kvm_vcpu_init { > >> + __u32 target; > >> + __u32 features[7]; > >> +}; > >> + > >> +struct kvm_sregs { > >> +}; > >> + > >> +struct kvm_fpu { > >> +}; > >> + > >> +struct kvm_guest_debug_arch { > >> +}; > >> + > >> +struct kvm_debug_exit_arch { > >> +}; > >> + > > > > This is a problem too: structure alignment is different > > in -m32 versus -m64 modes, which will affect the offset of > > the following fields. I think it's best to add a "padding" > > field in there, and size it to multiple of 8 bytes. > > Sorry, which field are you referring to? Sorry, I was wrong here: while it seems to be compler dependent gcc returns 0 as size of an empty structure. > > I think the same is a good idea for other empty stuctures, > > since otherwise the padding is implicit and > > not initialized by gcc, and it is hard not too leak > > info to userspace when you copy these structures out. > > > > And, it'll be handy if you want to extend the structures > > down the line. > > Sure, what would make some sense. > > Thanks, > > M. > -- > Fast, cheap, reliable. Pick two. -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
